I have found that Hub Watchdog is important for spotting slowdowns, but not for comparing absolute values.
I used to have one hub that controlled my zwave and zigbee devices (around 150).
I used to have Hub Watchdog go to a zwave plus device that was approximately 5 feet away to test response. Time was usually in the .2 -.3 range.
I then purchased a C7 and moved all my zwave devices to it. The physical position of the C7 was right next to the C5 (which now had only my zigbee devices). Times from Hub Watchdog was in the order of around .5-.7.
The cause of the increase was the route of the zwave mesh on the new C7. For whatever reason, on my C5 it was one hop away. On the C7 it was 3 hops away. Why? I really don't know.
So, the number of hops affect the speed as reported by Hub Watchdog.
I get all that. I'm mostly interested in comparing times across firmware revisions.
here's mine show me yours
i have been unable to figure out why i get an occasional slowdown.. it is not always at the same time.. i wonder if the hub cleanup runs at differnt times..
here are the times when it occured my checks are every hour .. so you can calculate the time from the last time shown above. They also may have been when i was running individual node repairs.. becuase that can cause an occasional hub busy message. That would not explain the zigbee slowdown however...
I'm on .148 on a C7 with zwave test device at maximum distance from hub. I was checking the routes from that side of the house. Ignore before 13:00 as I was doing a bunch of other stuff to the hub before then. I also don't currently have any zigbee test devices configured.
My hub started locking up daily and was trying to track it down. I think it was a misconfigured device in google home integration but hasn't been running long enough yet to confirm.
mine is not max distance it is avg. I was just checking on hub performance itself not trying to debug a certain location. The house will be empty for awhile and I want to be alerted if things stop working.
but With an extra hop I was seeing times similiar to yours. I am also c7 on .148
This is my Z-wave test switch I use so I'm not actually turning on and off lights all day long. It's a Honeywell (GE/Jasco) Z-Wave Plus Dimmer the same as most of the other switches in my house. Just need to wire up Line, Neutral, and Ground since I'm not actually controlling a load with it and that lets me use a regular 3 prong plug. I can move it around the house as I want to test different areas. I told all my kids repeatedly that it will kill them if they touch it and I think they finally got the point now to stay away from it. My daughter came to me last night and said: "The death plug is by the printer and I need to print something for school. Can you move it?" It was like 6 feet away from it but sure. I have a box ordered to put it in but it just hasn't gotten here yet. I haven't settled on a good test device for ZigBee yet since most of my ZigBee devices are battery-powered. Thinking about getting a ZigBee in-wall outlet.
Here are the tests from the next day. Pretty much the same results.
$2.09 plus tax could save a life. 
I have one of these on the way.
https://www.amazon.com/gp/product/B0765CZW22/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1
Funny I was looking at the same thing and thinking hmmmm then I saw your post 
I use that one for two receptacles I added in the attic for power tools.
i was thinking about your test device why not just use one of the plug in smart switches.. is there a reason you must use a wall switch?
I chose to use it since it is what all of my other devices are and it is an accurate representation of the radio strength of my devices. I also have a box of them lying around.
So, now that you have it working,
have you found significant areas of your house where the zwave signal is poor?
Are you using that box to test signal strength?
or speed of Hubitat execution?
or mesh issues?
Is there a distance factor that is significant (i.e. the farther from the hub, the slower the switch?)
(Sorry, for all the questions, I think that's a great idea!)
I did something like this last year but had several devices wired in such a fashion. Wanted a travelling "demo" - along with a C-5 hub,
I had a switch, a smart outlet in one gang box w/cord and another switch in another gangbox + regular outlet (for hub/wifi router). Then had some smart bulbs + buttons etc.
The one thing I failed to account for is the mesh getting out whack each time I shut everything down and set it up again in a different location. There was not enough time for things to stabilize and my first "demo" did not go so well - not working or reliable etc.
The other issue was I think one or two of the devices where z-wave not zw+. I might give it another shot sometime now that I am more familiar with potential issues.
Currently it's just running z-wave watchdog test so it's not turning on and off a light constantly. I only wired up Line, Ground, and Neutral. I use it for signal strength and routing tests. I also used it outside with an extension cord to test where devices could and could not reach. I also use it to test out rules for people in the forums so I'm not pissing off the wfie. The garage is by far my weakest area but my attic also has a few areas that give me problems. Distance can play a big factor in not only wireless speed but how many hops are required. It also was useful for testing areas I didn't want to attach a sensor to unless it would work in that area, such as a temperature sensor to the bottom of the tub on the outside.
We need the zwave routing hints or control.specified in the spec implemented.
I have a mimo2+ I just installed for a gate opener about 75 feet inside a battery case in back yd. It worked fine with older mimo+ on St.
I had a hell off a time getting it working on hubitat. Just putting it out there after pairing never worked. And it wouldn't pair in location but old one excluded fine on st. I even put an aeon ver 7 repeater right inside the slider. Using its test mode even at -3db i get consistant 10/10 succesfull test result. However, even after days and multiple repairs the mimo2+ will never use this repeater and is flaky at best.
Even to get it working at all I had to spend hours with it plugged into an extension cord after pairing indoors and slowly move it 10 feet at a time closer to final location with multiple repairs and tests . What a pain
I don't think my mailbox sensor which is even farther in front yd will ever work in hubitat
related .. how do you do an inclusion using the stick..
also i noticed an option to add a route.. maybe i can override the routing for the problem nodes has anyone looked into using that function on the stick?
Just checking in on outcome here... How is your setup running since putting all of your control apps on an isolated hub? Do your device hubs run any apps whatsoever? Also, are you running your control apps on the actual hub with the hubconnect controller app or completely different child hub? Are you using hubconnect http or socket method?
